source-pack-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@source-pack-mcpBuild a source pack on the effects of microplastics on human health"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
source-pack-mcp
An MCP server that produces a structured research source pack for a topic or question. Instead of an agent rummaging through search results and raw pages, it returns verified-at-source facts, quotes, key numbers, dates, primary links, and a coverage map of what's been said — as clean structured data.
What a source pack looks like
{
"pack_id": "uuid",
"query": "string",
"created_at": "ISO timestamp",
"sources": [
{
"url": "string",
"title": "string",
"domain": "string",
"retrieved_at": "ISO timestamp",
"facts": [{"fact": "string", "confidence": "high|medium|low"}],
"quotes": [{"text": "string", "context": "string"}],
"numbers": [{"value": "string", "unit": "string", "context": "string"}],
"dates": [{"date": "string", "event": "string"}],
"primary_links": ["string"]
}
],
"coverage_map": [
{"claim": "string", "mentioned_by": ["domain"], "contested_by": ["domain"], "status": "consistent|contested|isolated"}
],
"limitations": ["string"]
}Facts are extracted from actually-fetched sources, with the URL attached and a confidence label. The server never invents facts and never claims verification it didn't perform. coverage_map compares claims across sources: consistent, contested, or isolated.
Related MCP server: AutoWiki Agent
Tools
Tool | Purpose |
| Build a source pack for a query ( |
| Add a specific URL to an existing pack for extraction. |
| Retrieve a previously built pack by id. |
| List recent packs (id, query, created_at, source count). |
| Keyword search across previously built packs. |
Features
HTML → clean text via real DOM extraction (Mozilla Readability + Turndown), never regex parsing
Disk cache with TTL (default 1 hour)
robots.txt respect, rate-limited fetching (~1 req/sec)
Packs persisted as JSON in a configurable data dir (
.packs/)Streamable HTTP (default
127.0.0.1:3225) + stdio transportsNo API keys required, no auth needed for public content, no personal data
Quick start
Requirements: Node.js 22+
npm ci
npm run build
PORT=3225 node dist/src/http.jsConfigure with env vars (see .env.example):
Var | Default | Purpose |
| 8 | Max sources per pack |
| 15000 | Per-request fetch timeout |
| 1000 | Rate limit between fetches |
| unset | Optional JSON search API; falls back to HTML search |
| 3600 | Page cache TTL |
|
| Where packs are stored |
| 3225 | HTTP port |
Register in any MCP client:
{
"mcpServers": {
"source-pack": {
"command": "node",
"args": ["/path/to/source-pack-mcp/dist/src/http.js"],
"env": { "PORT": "3225" }
}
}
}Development
npm run check # typecheck + lint + build + test
npm test # run testsLicense
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-qualityCmaintenanceAn MCP server that provides fact-checking capabilities and truth anchoring for AI agents using verified data sources.Last updatedMIT
- Alicense-qualityCmaintenanceAn MCP server that compiles any text into a verifiable, graph-based knowledge base using deterministic chunking and parallel extraction of epistemology primitives.Last updatedMIT
- Alicense-qualityBmaintenanceThis MCP server enables AI agents to extract, verify, and cache factual claims from text using Claude Haiku and semantic fingerprinting, with tools for content verification, signal verification, and fact memory search.Last updated10MIT
- AlicenseAqualityBmaintenanceMCP server providing access to U.S. government primary-source records, fact-checks, news search, and trackers, with cross-referenced entity data and source links.Last updated477MIT
Related MCP Connectors
An MCP server for deep research or task groups
MCP server for generating rough-draft project plans from natural-language prompts.
Academic research MCP server for paper search, citation checks, graphs, and deep research.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/SarutobiSasuke8/source-pack-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server